Wildfire Simulation Research Articles

An integrated environmental analytics system (IDEAS) based on a DGGS

Discrete global grid systems (DGGS) have been proposed as a data model for a digital earth framework. We introduce a new data model and analytics system called IDEAS – integrated discrete environmental analysis system to create an operational DGGS-based GIS which is suitable for large scale environmental modelling and analysis. Our analysis demonstrates that DGGS-based GIS is feasible within a relational database environment incorporating common data analytics tools. Common GIS operations implemented in our DGGS data model outperformed the same operations computed using traditional geospatial data types. A case study into wildfire modelling demonstrates the capability for data integration and supporting big data geospatial analytics. These results indicate that DGGS data models have significant capability to solve some of the key outstanding problems related to geospatial data analytics, providing a common representation upon which fast and scalable algorithms can be built.

Predicting Emission Source Terms in a Reduced-Order Fire Spread Model—Part 1: Particulate Emissions

A simple, easy-to-evaluate, surrogate model was developed for predicting the particle emission source term in wildfire simulations. In creating this model, we conceptualized wildfire as a series of flamelets, and using this concept of flamelets, we developed a one-dimensional model to represent the structure of these flamelets which then could be used to simulate the evolution of a single flamelet. A previously developed soot model was executed within this flamelet simulation which could produce a particle size distribution. Executing this flamelet simulation 1200 times with varying conditions created a data set of emitted particle size distributions to which simple rational equations could be tuned to predict a particle emission factor, mean particle size, and standard deviation of particle sizes. These surrogate models (the rational equation) were implemented into a reduced-order fire spread model, QUIC-Fire. Using QUIC-Fire, an ensemble of simulations were executed for grassland fires, southeast U.S. conifer forests, and western mountain conifer forests. Resulting emission factors from this ensemble were compared against field data for these fire classes with promising results. Also shown is a predicted averaged resulting particle size distribution with the bulk of particles produced to be on the order of 1 μm in size.

Modelling the effect of accelerated forest management on long-term wildfire activity

We integrated a widely used forest growth and management model, the Forest Vegetation Simulator, with the FSim large wildfire simulator to study how management policies affected future wildfire over 50 years on a 1.3 million ha study area comprised of a US national forest and adjacent lands. The model leverages decades of research and development on the respective forest growth and wildfire simulation models, and their integration creates a strategic forest landscape model that has a high degree of transparency in the existing user communities. The study area has been targeted for forest restoration investments in response to wildland fires that are increasingly impacting ecological conditions, conservation areas, amenity values, and surrounding communities. We simulated three alternative spatial investment priorities and three levels of management intensity (area treated) over a 50-year timespan and measured the response in terms of area burned, fire severity, wildland-urban interface exposure and timber production. We found that the backlog of areas in need of restoration on the national forest could be eliminated in 20 years when the treatment rate was elevated to a maximum of 3× the current level. However, higher rates of treatments early in the simulation created a future need to address the rapid buildup of fuels associated with understory shrub and tree regeneration. Restoration treatments over time had a large effect on fire severity, on average reducing potential flame length by up to 26% for the study area within the first 20 years, whereas reductions in area burned were relatively small. Although wildfire contributed to reducing flame length over time, area burned was only 16% of the total disturbed area (managed and burned with prescribed fire) under the 3× management intensity. Interactions among spatial treatment scenarios and treatment intensities were minimal, although inter-annual variability was extreme, with the coefficient of variation in burned area exceeding 200%. We also observed simulated fires that exceeded four times the historically recorded fire size. Fire regime variability has manifold significance since very large fires can homogenize fuels and eliminate clumpy stand structure that historically reduced fire size and maintained landscape resiliency. We discuss specific research needs to better understand future wildfire activity and the relative influence of climate, fuels, fire feedbacks, and management to achieve fire resiliency goals on western US fire frequent forests.

Climate change effects on wildfire hazards in the wildland-urban-interface – Blue pine forests of Bhutan

Increased wildfire activity in the Himalayan Mountains due to climate change may place rural livelihoods at risk, yet potential climate change effects on forest fires in this region are poorly investigated. Here we use Bhutan’s blue pine (Pinus wallichiana) ecosystems to study the sensitivity of fire behavior to climatic changes. Wildland fires are one of the biggest threats to forest resources in Bhutan; blue pine ecosystems, in particular, are of high concern because of their importance for rural livelihoods and relatively high frequency of forest fires. Due to the geographical and socioeconomic characteristics of Bhutan, the region is highly sensitive to climate change. We investigated fire hazards in the wildland-urban-interface (WUI) of two valleys in Bhutan (Thimphu and Jakar), where human settlements and infrastructure are surrounded by blue pine forests. We applied FlamMap, a spatially-explicit wildfire simulation model, to simulate fire behavior under four climate scenarios. As indicators of fire behavior, we used flame length, rate of spread, crown fire activity, burn probability, and fire size. With the simulation outcomes we constructed a fire hazard map showing the hotspots of forest fire susceptibility. FlamMap predicts a two-fold increase in fire hazards in the WUI for both study areas owing to climatic changes. The capital city of Thimphu has, on average, greater fire hazards than Jakar, though fire hazards are spatially variable within both study areas. Our study contributes to the understanding of and ability to predict forest fire hazards in Himalayan blue pine ecosystems. The findings will help to more precisely allocate fire management resources in the WUI, plan suburban development to minimize fire risk to livelihoods, and adapt forest management in the face of climate change.

Ladder Fuel Treatments Effect Burn Area of Forest Fires in Semi-Arid High Elevation Climates

Forest fires destroy millions of acres, threaten populations, and are becoming increasingly severe. Efforts to reduce the starting of wildfires has led to increasing quantities of ladder fuels. Ladder fuels are any fuels that create a vertical continuity in a forest. When a fire does occur, these ladder fuels promote the spread of fires to the forest crown. We hypothesized that if the density of ladder fuel is reduced, the burn area in a semi-arid high-altitude forest will decrease. To test the hypothesis, a series of computer simulations were run testing 50 random ignition sites in a representative semi-arid high elevation forest with different levels of ladder fuel treatment using the FlameMap wildfire simulator with US Forest Service landscape and weather data. There was a large variation in the extent of the fires based on the ignition location, ranging from 13 to 22,506 acres. The study showed a statistically significant reduction in average total burn area after 10 days when 5% or more of the total acreage had been treated (p < 0.01). There were further reductions in burn area with increasing ladder fuel treatment from a 9% reduction with 5% treatment to a 51% reduction with 20% treatment (p < 0.001). This study supports the effectiveness of ladder fuel treatments in reducing burn area.

Thermo- and physicochemical properties of native and exotic forest species of Valparaíso, Chile, as essential information for fire risk management

Wildfires in the Valparaíso region (Chile) account for one of the main threats to local biodiversity, ecosystem services and infrastructure. This study focused on producing an initial record of thermo- and physicochemical properties of local forest species. For this purpose, leaf samples of species found in the Peñuelas Lake National Reserve, namely Pinus radiata, Eucalyptus globulus, Acacia dealbata, Quillaja saponaria and Cryptocarya alba, were collected and analysed. Higher and lower heating value, flash point, density and moisture content tests were performed for each sample. Overall results showed that lower heating values measured for both native and exotic species could indicate a high energy release source in wildfires. However, differences in the flash point between species indicated that C. alba and E. globulus had a lower ignition resistance than other species tested, possibly due to a lower flash point. In contrast, Q. saponaria and A. dealbata had the highest flash point for native and exotic species respectively. Finally, all presented data and procedures were aimed at establishing a foundation for a national database of critical forest species properties to be used in wildfire simulation tools. This database will enhance forest fire management effectiveness in Chile.

Wildfire Detection Probability of MODIS Fire Products under the Constraint of Environmental Factors: A Study Based on Confirmed Ground Wildfire Records

The Moderate Resolution Imaging Spectroradiometer (MODIS) has been widely used for wildfire occurrence and distribution detecting and fire risk assessments. Compared with its commission error, the omission error of MODIS wildfire detection has been revealed as a much more challenging, unsolved issue, and ground-level environmental factors influencing the detection capacity are also variable. This study compared the multiple MODIS fire products and the records of ground wildfire investigations during December 2002–November 2015 in Yunnan Province, Southwest China, in an attempt to reveal the difference in the spatiotemporal patterns of regional wildfire detected by the two approaches, to estimate the omission error of MODIS fire products based on confirmed ground wildfire records, and to explore how instantaneous and local environmental factors influenced the wildfire detection probability of MODIS. The results indicated that across the province, the total number of wildfire events recorded by MODIS was at least twice as many as that in the ground records, while the wildfire distribution patterns revealed by the two approaches were inconsistent. For the 5145 confirmed ground records, however, only 11.10% of them could be detected using multiple MODIS fire products (i.e., MOD14A1, MYD14A1, and MCD64A1). Opposing trends during the studied period were found between the yearly occurrence of ground-based wildfire records and the corresponding proportion detected by MODIS. Moreover, the wildfire detection proportion by MODIS was 11.36% in forest, 9.58% in shrubs, and 5.56% in grassland, respectively. Random forest modeling suggested that fire size was a primary limiting factor for MODIS fire detecting capacity, where a small fire size could likely result in MODIS omission errors at a threshold of 1 ha, while MODIS had a 50% probability of detecting a wildfire whose size was at least 18 ha. Aside from fire size, the wildfire detection probability of MODIS was also markedly influenced by weather factors, especially the daily relative humidity and the daily wind speed, and the altitude of wildfire occurrence. Considering the environmental factors’ contribution to the omission error in MODIS wildfire detection, we emphasized the importance of attention to the local conditions as well as ground inspection in practical wildfire monitoring and management and global wildfire simulations.

FIRE MODELLING TO ASSESS SPATIAL PATTERNS OF WILDFIRE EXPOSURE IN ARDABIL, NW IRAN

Abstract. Fire exposure describes the spatial juxtaposition of values with fire behaviour in terms of likelihood and intensity. Wildfire exposure analysis is based on the estimation of the potential wildfire intensity and on the burn probability. Fire modelling can produce spatially explicit information on fire spread and behaviour, and offers a feasible method to simulate, map, and analyse fire exposure. FlamMap Minimum Travel Time (MTT) algorithm (Finney, 2006) was used to conduct wildfire simulations considering historical data of fuel moisture conditions and winds, as well as the most frequent wind directions and historical ignition locations (2005-2018). Analysis was conducted on spatial and quantitative variations in selected fire hazard and exposure factors, namely Burn Probability (BP), Conditional Flame Length (CFL) and Fire Size (F). We observed pronounced spatial variations among and between municipalities in the factors, especially for those in the northern and southern parts of Ardabil. The variations across the burnable area of the municipalities can be fundamentally related to a number of factors, including spatial variation in ignition locations, fuel moisture and load, weather conditions, and topography of the terrain. The findings can provide information and support in wildfire management planning and fire risk mitigation activities.

Wildfire exposure to the wildland urban interface in the western US

Predicting wildfire disasters presents a major challenge to the field of risk science, especially when fires propagate long distances through diverse fuel types and complex terrain. A good example is in the western US where large tracts of public lands routinely experience large fires that spread from remote wildlands into developed areas and cause structure loss and fatalities. In this paper we provide the first comprehensive assessment of where public wildlands potentially contribute wildfire exposure to communities in the 11 western US states. We used simulation modeling to map and characterize the composition of the source landscapes (firesheds) and recipient communities in terms of fuels, fire behavior and forest management suitability. The information was used to build a prototype investment prioritization framework that targets highly exposed communities where forest and fuel management activities are feasible. We found that simulated wildfires ignited on national forests can potentially affect about half of the communities in the western US (2560 out of 5118), with 90% of exposure affecting the top 20% of the communities (n = 516). Firesheds within national forests, defined as areas that have the potential to expose communities to fire, were estimated at 35 million ha (62% of the total national forest area), and were almost three times larger than the affected community lands. Large contiguous areas of wildfire transmission were evident on a number of national forests. Only 22% of the fireshed area is forested, fire-adapted, and lies within land management designations that allow mechanical fuels management. The methods demonstrate how cross-boundary exposure can be factored into prioritizing federal investments in hazardous fuels reduction on national forests in concert with community protection measures. The results can also help scale wildfire governance systems to match the geography of risk from large wildfire events, which augments existing assessments that do not explicitly identify the source of risk to communities.

Giving life to the map can save more lives. Wildfire scenario with interoperable simulations

Abstract. In the Mediterranean region, drier and hotter summers are leading to more likely and severe wildfires. The authors propose an innovative approach for situational awareness by giving life to maps and exploiting interoperable GIS, hazard models, simulations, and interconnection analysis processes aimed to enhance preparedness and strengthen the resilience of responding organizations. The information related to a virtual city and its countryside has been implemented in the terrain of simulation systems. The TIGER wildfire model software has been adapted to a scenario where districts, refugee camps and critical infrastructures can be impacted by a fire and has been linked to a smoke dispersion model, and associated impacts to the electricity network and roads. The transfer of computed fire propagation and combustion data to the AI-powered SWORD simulation enable more accurate computing of damage and loss. In SWORD, civil protection, military assets and humanitarian actions can be performed for training and operation preparation. The simulation data about fire and assets’ deployments can feed a web app map or a command and control system, thus providing situational awareness for optimal decision-making, and analysis about people in danger, network interconnections and potential service disruption. Disaster managers and commanders can interact with simulated assets performing their chosen courses of action and analyse the outcomes.In conclusion, tests in a wildfire case study demonstrated a high level of interoperability among those systems and the possibility to provide updated situational awareness leading to better emergency preparedness and critical infrastructure resilience building, finally contributing to save more lives.

A review of a new generation of wildfire–atmosphere modeling

One of the first significant developments in wildfire modeling research was to introduce heat flux as wildfire line intensity (kW·m–1). This idea could be adapted to using weather station measureme...

Long-Term Impacts of Fuel Treatment Placement with Respect to Forest Cover Type on Potential Fire Behavior across a Mountainous Landscape

Research Highlights: The impact of variation in fuels and fuel dynamics among forest cover types on the outcome of fuel treatments is poorly understood. This study investigated the potential effects of treatment placement with respect to cover type on the development of potential fire behavior over time for 48 km2 of forest in Colorado, USA. Our findings can inform the placement of fuel treatments in similar forests to maximize their effectiveness and longevity. Background and Objectives: Efficient placement of fuel treatments is essential to maximize the impact of limited resources for fuels management. We investigated how the placement of treatments with respect to forest cover type affected the rate of spread, size, and prevalence of different fire types for simulated wildfires for 50 years after treatment. Materials and Methods: We generated an analysis landscape consisting of two cover types: stands on southerly aspects had low rates of tree growth and regeneration compared to stands on northerly aspects. We then simulated 1) thinning treatments across 20% of the landscape, with treatments exclusively located on either southerly (‘south treatment’) or northerly (‘north treatment’) aspects; 2) subsequent tree growth and regeneration; and 3) wildfires at 10-year intervals. Finally, we used metrics of fuel hazard and potential fire behavior to understand the interplay between stand-level fuel dynamics and related impacts to potential fire behavior across the broader landscape. Results: Although post-treatment metrics of stand-level fuel hazard were similar among treatment scenarios, only the south treatment reduced rates of fire spread and fire size relative to no treatment. Most differences in modeled fire behavior between treatment scenarios disappeared after two decades, despite persistently greater rates of stand-level fuel hazard development post-treatment for the north treatment. For all scenarios, the overall trajectory was of shrinking fires and less crown fire behavior over time, owing to crown recession in untreated stands. Conclusions: Systematic differences among cover types, such as those in our study area, have the potential to influence fuel treatment outcomes. However, complex interactions between treatment effects, topography, and vegetation structure and dynamics warrant additional study.

A Data-Driven Fire Spread Simulator: Validation in Vall-llobrega's Fire

While full-physics fire models continue to be unsuitable for wildfire emergency situations, the so-called operational fire spread simulators are incapable of providing accurate estimations of the macroscopic fire behaviour while quickly reacting to a change of governing spread mechanisms. A promising approach to overcome these limitations are data-driven simulators, which assimilate observed data with the aim of improving their forecast with affordable computation times. Although preliminary results obtained by several data-driven simulators are promising, this scheme needs intensive validation. Detailed studies of the particular aspects related to data assimilation are essential to gain insight about the applicability of this approach to operational wildfire simulation. This paper presents the validation of the simulator presented in \citep{Rios2014a, Rios2016, Rios2018} with a large scenario of real complexity with intricate terrain. The study case corresponds to a wildfire of significant repercussions occurred in Catalonia in March 2014. We employed as reference data the event reconstruction performed by the Catalan Fire Service and validated with operational observations. Detailed information about fuel and meteorology was collected by the fire brigades and allowed reconstructing the fire development with Farsite, a widely employed simulator. Subsequently, our simulator was tested without a detailed description of the fuel and wind parameters, i.e. imitating its intended deployment conditions. It proved capable of automatically estimating them and correctly simulating the fire spread. Additionally, the effect of the assimilation window on the forecast accuracy was analysed. These results showed that the simulator is able to correctly handle complex terrain and wind situations to successfully deliver a short-term fire-front forecast in those real and complex scenarios.

Predictive modeling of wildfires: A new dataset and machine learning approach

Wildfires, whether natural or caused by humans, are considered among the most dangerous and devastating disasters around the world. Their complexity comes from the fact that they are hard to predict, hard to extinguish and cause enormous financial losses. To address this issue, many research efforts have been conducted in order to monitor, predict and prevent wildfires using several Artificial Intelligence techniques and strategies such as Big Data, Machine Learning, and Remote Sensing. The latter offers a rich source of satellite images, from which we can retrieve a huge amount of data that can be used to monitor wildfires. The method used in this paper combines Big Data, Remote Sensing and Data Mining algorithms (Artificial Neural Network and SVM) to process data collected from satellite images over large areas and extract insights from them to predict the occurrence of wildfires and avoid such disasters. For this reason, we implemented a methodology that serves this purpose by building a dataset based on Remote Sensing data related to the state of the crops (NDVI), meteorological conditions (LST), as well as the fire indicator “Thermal Anomalies”, these data, were acquired from “MODIS” (Moderate Resolution Imaging Spectroradiometer), a key instrument aboard the Terra and Aqua satellites. This dataset is available on GitHub via this link (https://github.com/ouladsayadyounes/Wildfires). Experiments were made using the big data platform “Databricks”. Experimental results gave high prediction accuracy (98.32%). These results were assessed using several validation strategies (e.g., classification metrics, cross-validation, and regularization) as well as a comparison with some wildfire early warning systems.

Modelling of the Radiant Heat Flux and Rate of Spread of Wildfire within the Urban Environment

One approach to increase community resilience to wildfire impacts is the enhancement of residential construction standards in an effort to provide protective shelters for families within their own homes. Current wildfire models reviewed in this study assume fire growth is unrestricted by vegetation fuel bed geometry; the head fire has attained a quasi-steady rate of spread; and the shielding effects of urban development are ignored. As a result, radiant heat flux may be significantly overestimated for small vegetation fires in road reserves, urban parklands, and similar scenarios. This paper proposes two new models to address this issue, and utilises two case studies for comparison against existing approaches. The findings are significant as this is the first study to analyse these factors from a fire engineering perspective, and to demonstrate that the use of landscape scale or siege wildfire models may not be appropriate within the urban context. The development of enhanced wildfire models will have a significant impact on town planning and construction practices in areas prone to wildfires, as well as firefighting suppression efforts when these events occur.

Calculation of Critical Water Flow Rates for Wildfire Suppression

Predicting water suppression requirements and its impacts on firefighting strategies and logistics within the urban environment has been the subject of many previous studies, however the same level of research has yet to be applied in the realm of wildfire suppression. To work towards addressing this knowledge gap, this paper provides guidance for Incident Controllers in relation to critical water flow rates required to extinguish large wildfire across a wide range of forest fuel loads, fire weather and active fire front depths. This is achieved through mathematical empirical analysis of water flow rates required for head fire suppression during 540 simulated wildfires in forest vegetation. This research applies a fire engineering approach to wildfire suppression logistics and deterministically assess the suitability of appliance and aircraft based head fire suppression. The results highlight the limitations of offensive wildfire suppression involving direct head fire attacks by appliances once wildfires attain a quasi-steady state in forest fuels.

Analysis of the uncertainty of fuel model parameters in wildland fire modelling of a boreal forest in north-east China

Fire propagation is inevitably affected by fuel-model parameters during wildfire simulations and the uncertainty of the fuel-model parameters makes forecasting accurate fire behaviour very difficult. In this study, three different methods (Morris screening, first-order analysis and the Monte Carlo method) were used to analyse the uncertainty of fuel-model parameters with FARSITE model. The results of the uncertainty analysis showed that only a few fuel-model parameters markedly influenced the uncertainty of the model outputs, and many of the fuel-model parameters had little or no effect. The fire-spread rate is the driving force behind the uncertainty of other fire behaviours. Thus, the highly uncertain fuel-model parameters associated with spread rate should be used cautiously in wildfire simulations. Monte Carlo results indicated that the relationship between model input and output was non-linear and neglecting fuel-model parameter uncertainty of the model would magnify fire behaviours. Additionally, fuel-model parameters have high input uncertainty. Therefore, fuel-model parameters must be calibrated against actual fires. The highly uncertain fuel-model parameters with high spatial-temporal variability consisted of fuel-bed depth, live-shrub loading and 1-h time-lag loading are preferentially chosen as parameters to calibrate several wildfires.

A web-based wildfire simulator for operational applications

Wildfire simulators and decision support systems can assist the incident command teams in charge of tactical wildfire suppression. This paper presents a web-based wildfire simulator developed to provide real-time support for wildfire management. The paper describes the overall software architecture, the modelling chain characteristics and the results produced by the simulator considering a set of actual wildfires that occurred in the island of Sardinia, Italy. The simulator consists of a graphical user interface that deals with data input–output management, a mass-consistent model devoted to the downscaling of wind fields, and a module that provides a spatially explicit representation of wildfire propagation. The simulator is a client‐server application that is operated through a web-based graphical user interface that leaves the computational work to a dedicated server; most of the code is parallelised in order to minimise computational run-time. The validation phase demonstrated the capabilities of the simulator in providing wildfire predictions with a substantial agreement with actual wildfires, and a computational cost suitable for faster than real-time applications. The simulator is proposed as a tool to provide assistance to civil protection and fire management agencies during the incident response phase. The simulator is also appropriate for the training of personnel.

Kalman Filter-Based Large-Scale Wildfire Monitoring With a System of UAVs

Wildfires play an important role in forest management, while an accurate assessment of current wildfire status is imperative for fire management. In this paper, a Kalman filter-based methodology is proposed to estimate the wildfire progress with online measurements that are collected by a system of unmanned aerial vehicles. The method is developed to estimate the wildfire propagation behavior as well as fire front contour. It is enabled by developing a scalar field wildfire model and utilizing the Kalman filter to estimate the parameters of the model. In addition, an uncertainty function is developed to describe the performance of the estimation results. The effectiveness of the algorithm is demonstrated in an independent fire simulation environment.

Coupled Atmosphere‐Fire Simulations of the Black Saturday Kilmore East Wildfires With the Unified Model

AbstractA model for the spread of a wildfire is developed within the U.K. Met Office Unified Model (UM) and used to simulate the Kilmore East fire complex (in southeastern Australia) on Black Saturday (7 February 2009). The UM is configured with four nests with horizontal grid spacings of 4 km, 1.5 km, 444 m, and 144 m. In the first simulation, the UM simply provides predictions of the near‐surface conditions for the wildfire model with no feedbacks to the atmosphere from the fire. In the second, the atmosphere and fire are coupled, allowing the fire to affect the local‐scale weather. The agreement between the coupled simulation and the observed fire behavior is reasonably good. For example, the area burnt is approximately 80% of the actual area burnt. However, such agreement is achieved only by coupling the fire to the atmosphere and, importantly, by igniting 18 additional fires at the times and places the long‐range transport of burning material (spotting) was observed. Without coupling the burnt area is about half of that observed. The calculations reported here suggest that the behavior and spread of fires like the Kilmore East fire are predictable but only when long‐range spotting is included and the atmosphere and fires are coupled. Additional numerical experiments with coarser grids suggest that, although the details of the fire spread are lost, a grid spacing of 1.5 km may be sufficient to simulate the main features of the fire spread.